techniques are required, we turned our attention to fluorous
7
chemistry as a promising option for catalyst recovery. This
Scheme 1. Synthesis of Fluorous Prolinol 7
recently introduced methodology is an alternative solution-
phase tagging approach in catalysis and high-throughtput
synthesis. Highly hydrophobic perfluoroalkyl phase-tags,
8
9
instead of polymers, are used to facilitate the separation of
the fluorous compounds from a reaction mixture using
fluorous liquid-liquid or solid-liquid extraction.
Inspired by the achievements of the fluorous chemistry,
we reasoned that it might be possible to prepare the fluorous
analogue of CBS catalyst (1a or 1b) and employ it in
enantioselective reduction of ketones. To minimize the
interference with the catalyst active site, we envisaged a
fluorous analogue that has the perfluorinated tags in the para
position of the phenyl residue.
We built upon the sequence described by Kanth and
10
Periasamy to synthesize 1 and applied similar methodology
for the preparation of 7. Moreover, our synthetic path is
basically the same as the one Bolm and Kim developed
1
1
recently for the synthesis of fluorous prolinol derivatives.
We converted L-proline (2) into its N-protected methyl
ester derivative 3 in a one-pot procedure, followed by
addition of 4-bromophenylmagnesium bromide to afford 4.
Subsequent treatment with NaOH gave cyclic carbamate 5
quantitatively. To access the fluorous derivative 6, we made
use of the Ullmann-type reaction between perfluorooctyl
iodide and cyclic carbamate 5 in the presence of freshly
activated copper powder. Contrary to a recent report, this
coupling step worked well in our hands, providing easy
access to fluorous proline derivative 6. Finally, the hydrolysis
of 6 gave the free amino alcohol 7 in an excellent yield. In
performing asymmetric reduction with diphenyl prolinol 7,
it is essential to prepare and use its oxazaborolidine derivative
prior to the catalytic application (Figure 1).
complex in tetrahydrofuran at room temperature. The result
showed that fluorous catalyst 8a perfomed as efficiently as
the original catalyst 1b (Table 1, entry 1). Once a reduction
was complete, it was worked up using fluorous SPE
methodology. Unfortunately, the catalyst 8a did not remain
intact under the recycling conditions, hydrolysis of the
oxazaborolidine catalyst to fluorous prolinol 7 occurred.
Although we were glad that the catalyst 8a performed as
well as the original catalyst 1b and that it could be easily
separated from the reaction products, the recovery of this
fluorous catalyst is not viable in this way because of its
hydrolysis.
12
1
1
As the above protocol returned the catalyst 8a in the form
of fluorous prolinol 7, we reasoned that the utilization of in
situ-generated oxazaborolidine might be the key to the
development of recoverable CBS reduction methodology.
Furthermore, the in situ formation of catalyst eliminates the
necessity of isolating the catalyst as a discrete step, providing
a simple method for asymmetric ketone reduction. The in
situ preparation and usage of chiral oxazaborolidines is a
Following established procedures, the fluorous analogue
8
a was formed by the addition of trimethylboroxine to a
1
solution of 7 in toluene. However, NMR analysis ( H and
DOSY experiments) of the oxazaborolidine product 8a
indicated the formation of unidentified fluorous impurities.
This catalyst, although not pure, was tested in asymmetric
reduction, since our aim was to determine whether it was
catalytically efficient and also establish the catalyst endurance
under recycling conditions. To comparatively evaluate our
1
3
known procedure that was employed successfully even on
1
4
an industrial scale.
catalyst, acetophenone (9) was reduced with BH ‚THF
3
To demonstrate the ability of fluorous prolinol 7 to serve
as a catalyst precursor, first we generated the B-OMe
(
6) Giffels, G.; Beliczey, J.; Felder, M.; Kragl, U. Tetrahedron: Asym-
metry 1998, 9, 691.
7) (a) Curran, D. P. In Stimulating Concepts in Chemistry; Stoddard,
F., Reinhoud, D., Shibasaki, M., Eds.; Wiley-VCH: New York, 2000; p
5. (b) Gladysz, J. A.; Curran, D. P. Tetrahedron 2002, 58, 3823 and the
(
2
following articles in this special issue entitled “Fluorous Chemistry”. (c)
The Handbook on Fluorous Chemistry; Gladysz, J. A., Horv a´ th, I. T.,
Curran, D. P., Eds.; Wiley-VCH: New York, 2004.
(
8) Horv a´ th, I. T.; R a´ bai, J. Science 1994, 266, 72.
(9) Studer, A.; Hadida, S.; Ferritto, R.; Kim, S.-Y. Jeger, P.; Wipf, P.;
Curran, D. P. Science 1997, 275, 823.
(
(
10) Kanth, J. V. B.; Periasamy, M. Tetrahedron 1993, 49, 5127.
11) Park, J. K.; Lee, H. G.; Bolm, C.; Kim, B. M. Chem. Eur. J. 2005,
1
1, 945. During the preparation of this manuscript, we became aware of
the presence of this recent paper closely related to our work. Compounds
, 5, and 6 are also intermediates in their synthetic path. However, their
4
target was a fluorous N-Me diphenylprolinol derivative that was used in
asymmetric diethyl- and diphenylzinc addition.
Figure 1. Fluorous oxazaborolidine catalysts.
3244
Org. Lett., Vol. 7, No. 15, 2005